Qualitative Analysis I: Principles and Manual Coding
❦
What Analysis Is
Qualitative analysis is an iterative, interpretive process of making sense of data through organizing, coding, comparing, and theorizing. It is not mechanical, and it begins during data collection. Miles et al. (2020) describe three concurrent activities: data condensation (selecting, focusing, and simplifying), data display (organizing information in matrices, networks, and charts), and conclusion drawing and verification.
Coding
A code is a short label that captures the meaning of a segment of data. Saldaña (2021) describes many types of code and two main "cycles":
| Cycle | Purpose | Examples of coding methods |
|---|---|---|
| First cycle | Initial labeling of data | Descriptive (topic), In Vivo (participants' own words), process (actions, using gerunds), emotion, values, versus (conflicts), initial or open coding |
| Second cycle | Grouping and conceptualizing | Pattern coding, focused coding, axial coding, theoretical coding |
Types of coding approach
- Inductive (data-driven): codes arise from the data.
- Deductive (theory-driven): codes come from a framework.
- Abductive or hybrid: a combination.
- Semantic codes capture explicit meaning; latent codes capture underlying ideas, assumptions, and concepts (Braun & Clarke, 2006).
Steps in Manual Coding
1. Prepare transcripts with wide margins and line numbers.
2. Read the whole transcript first.
3. Code line by line or segment by segment, using short labels; code everything of relevance to your questions, and allow several codes for one segment.
4. Keep a codebook: for each code, record its name, definition, example, and notes on when to use it.
5. Compare segments and codes: merge, split, and rename.
6. Group codes into categories and then into themes or higher-order concepts.
7. Write memos about patterns.
8. Look for negative cases and variation.
9. Display the results in tables and maps.
10. Return to the data to check that interpretations are grounded.
Manual Methods and Tools
| Tool | How it is used |
|---|---|
| Paper and highlighters | Print transcripts; mark segments; cut and sort into piles (the "cut and paste" method) |
| Word processor | Use comments, highlighting, and tables; copy segments into theme documents |
| Spreadsheet (Excel or Google Sheets) | Columns for participant, segment, code, category, theme, and notes; filter and sort |
| Index cards or sticky notes | Arrange and rearrange codes into themes on a wall |
| Mind maps and whiteboards | Explore relationships |
Manual analysis suits small datasets and helps students to stay close to the data. For larger projects, software helps to manage and retrieve (Chapter 15).
A Worked Example (Illustrative Data)
Data extract (participant P4, a woman who runs a small grocery shop): "At first I was afraid. If the machine failed, who would help me? But my daughter showed me, and now the customers prefer it. I do not have to count so much cash."
| Segment | Initial code | Category | Possible theme |
|---|---|---|---|
| "At first I was afraid" | Fear of the new | Initial apprehension | Moving from fear to confidence |
| "If the machine failed, who would help me?" | Worry about support | Need for assistance | Dependence on support networks |
| "My daughter showed me" | Learning from family | Social support | Dependence on support networks |
| "Customers prefer it" | Customer demand | Perceived benefits | Reasons for continuing |
| "I do not have to count so much cash" | Reduced cash handling | Perceived benefits | Reasons for continuing |
Across participants, similar codes are grouped, and the researcher develops themes that are patterns of meaning organized around a central idea, not just topics.
A Codebook Entry (Example)
| Code | Definition | When to use | Example |
|---|---|---|---|
| Learning from family | Participant describes learning or being helped by a family member | Any mention of family members teaching, helping, or encouraging use | "My daughter showed me" |
Handling Variation and Negative Cases
Look for cases that contradict your emerging pattern and explain them, or refine the pattern. A theme that fits all cases without exception may be too general.
Common Mistakes
- Coding only the first few transcripts thoroughly.
- Creating too many codes without grouping.
- Producing a list of topics instead of themes.
- Not keeping a codebook or audit trail.
- Assuming that frequency equals importance.
CHAPTER 14